qiskit 1.3.0b1__cp39-abi3-macosx_10_9_x86_64.whl

qiskit/primitives/backend_sampler_v2.py

@@ -0,0 +1,296 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2024.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+
+"""Sampler V2 implementation for an arbitrary Backend object."""
+
+from __future__ import annotations
+
+import warnings
+from collections import defaultdict
+from dataclasses import dataclass
+from typing import Iterable
+
+import numpy as np
+from numpy.typing import NDArray
+
+from qiskit.circuit import QuantumCircuit
+from qiskit.primitives.backend_estimator import _run_circuits
+from qiskit.primitives.base import BaseSamplerV2
+from qiskit.primitives.containers import (
+    BitArray,
+    DataBin,
+    PrimitiveResult,
+    SamplerPubLike,
+    SamplerPubResult,
+)
+from qiskit.primitives.containers.bit_array import _min_num_bytes
+from qiskit.primitives.containers.sampler_pub import SamplerPub
+from qiskit.primitives.primitive_job import PrimitiveJob
+from qiskit.providers.backend import BackendV1, BackendV2
+from qiskit.result import Result
+
+
+@dataclass
+class Options:
+    """Options for :class:`~.BackendSamplerV2`"""
+
+    default_shots: int = 1024
+    """The default shots to use if none are specified in :meth:`~.run`.
+    Default: 1024.
+    """
+
+    seed_simulator: int | None = None
+    """The seed to use in the simulator. If None, a random seed will be used.
+    Default: None.
+    """
+
+
+@dataclass
+class _MeasureInfo:
+    creg_name: str
+    num_bits: int
+    num_bytes: int
+    start: int
+
+
+class BackendSamplerV2(BaseSamplerV2):
+    """Evaluates bitstrings for provided quantum circuits
+
+    The :class:`~.BackendSamplerV2` class is a generic implementation of the
+    :class:`~.BaseSamplerV2` interface that is used to wrap a :class:`~.BackendV2`
+    (or :class:`~.BackendV1`) object in the class :class:`~.BaseSamplerV2` API. It
+    facilitates using backends that do not provide a native
+    :class:`~.BaseSamplerV2` implementation in places that work with
+    :class:`~.BaseSamplerV2`. However,
+    if you're using a provider that has a native implementation of
+    :class:`~.BaseSamplerV2`, it is a better choice to leverage that native
+    implementation as it will likely include additional optimizations and be
+    a more efficient implementation. The generic nature of this class
+    precludes doing any provider- or backend-specific optimizations.
+
+    This class does not perform any measurement or gate mitigation.
+
+    Each tuple of ``(circuit, <optional> parameter values, <optional> shots)``, called a sampler
+    primitive unified bloc (PUB), produces its own array-valued result. The :meth:`~run` method can
+    be given many pubs at once.
+
+    The options for :class:`~.BackendSamplerV2` consist of the following items.
+
+    * ``default_shots``: The default shots to use if none are specified in :meth:`~run`.
+      Default: 1024.
+
+    * ``seed_simulator``: The seed to use in the simulator. If None, a random seed will be used.
+      Default: None.
+
+    .. note::
+
+        This class requires a backend that supports ``memory`` option.
+
+    """
+
+    def __init__(
+        self,
+        *,
+        backend: BackendV1 | BackendV2,
+        options: dict | None = None,
+    ):
+        """
+        Args:
+            backend: The backend to run the primitive on.
+            options: The options to control the default shots (``default_shots``) and
+                the random seed for the simulator (``seed_simulator``).
+        """
+        self._backend = backend
+        self._options = Options(**options) if options else Options()
+
+    @property
+    def backend(self) -> BackendV1 | BackendV2:
+        """Returns the backend which this sampler object based on."""
+        return self._backend
+
+    @property
+    def options(self) -> Options:
+        """Return the options"""
+        return self._options
+
+    def run(
+        self, pubs: Iterable[SamplerPubLike], *, shots: int | None = None
+    ) -> PrimitiveJob[PrimitiveResult[SamplerPubResult]]:
+        if shots is None:
+            shots = self._options.default_shots
+        coerced_pubs = [SamplerPub.coerce(pub, shots) for pub in pubs]
+        self._validate_pubs(coerced_pubs)
+        job = PrimitiveJob(self._run, coerced_pubs)
+        job._submit()
+        return job
+
+    def _validate_pubs(self, pubs: list[SamplerPub]):
+        for i, pub in enumerate(pubs):
+            if len(pub.circuit.cregs) == 0:
+                warnings.warn(
+                    f"The {i}-th pub's circuit has no output classical registers and so the result "
+                    "will be empty. Did you mean to add measurement instructions?",
+                    UserWarning,
+                )
+
+    def _run(self, pubs: list[SamplerPub]) -> PrimitiveResult[SamplerPubResult]:
+        pub_dict = defaultdict(list)
+        # consolidate pubs with the same number of shots
+        for i, pub in enumerate(pubs):
+            pub_dict[pub.shots].append(i)
+
+        results = [None] * len(pubs)
+        for shots, lst in pub_dict.items():
+            # run pubs with the same number of shots at once
+            pub_results = self._run_pubs([pubs[i] for i in lst], shots)
+            # reconstruct the result of pubs
+            for i, pub_result in zip(lst, pub_results):
+                results[i] = pub_result
+        return PrimitiveResult(results, metadata={"version": 2})
+
+    def _run_pubs(self, pubs: list[SamplerPub], shots: int) -> list[SamplerPubResult]:
+        """Compute results for pubs that all require the same value of ``shots``."""
+        # prepare circuits
+        bound_circuits = [pub.parameter_values.bind_all(pub.circuit) for pub in pubs]
+        flatten_circuits = []
+        for circuits in bound_circuits:
+            flatten_circuits.extend(np.ravel(circuits).tolist())
+
+        # run circuits
+        results, _ = _run_circuits(
+            flatten_circuits,
+            self._backend,
+            memory=True,
+            shots=shots,
+            seed_simulator=self._options.seed_simulator,
+        )
+        result_memory = _prepare_memory(results)
+
+        # pack memory to an ndarray of uint8
+        results = []
+        start = 0
+        for pub, bound in zip(pubs, bound_circuits):
+            meas_info, max_num_bytes = _analyze_circuit(pub.circuit)
+            end = start + bound.size
+            results.append(
+                self._postprocess_pub(
+                    result_memory[start:end],
+                    shots,
+                    bound.shape,
+                    meas_info,
+                    max_num_bytes,
+                    pub.circuit.metadata,
+                )
+            )
+            start = end
+
+        return results
+
+    def _postprocess_pub(
+        self,
+        result_memory: list[list[str]],
+        shots: int,
+        shape: tuple[int, ...],
+        meas_info: list[_MeasureInfo],
+        max_num_bytes: int,
+        circuit_metadata: dict,
+    ) -> SamplerPubResult:
+        """Converts the memory data into an array of bit arrays with the shape of the pub."""
+        arrays = {
+            item.creg_name: np.zeros(shape + (shots, item.num_bytes), dtype=np.uint8)
+            for item in meas_info
+        }
+        memory_array = _memory_array(result_memory, max_num_bytes)
+
+        for samples, index in zip(memory_array, np.ndindex(*shape)):
+            for item in meas_info:
+                ary = _samples_to_packed_array(samples, item.num_bits, item.start)
+                arrays[item.creg_name][index] = ary
+
+        meas = {
+            item.creg_name: BitArray(arrays[item.creg_name], item.num_bits) for item in meas_info
+        }
+        return SamplerPubResult(
+            DataBin(**meas, shape=shape),
+            metadata={"shots": shots, "circuit_metadata": circuit_metadata},
+        )
+
+
+def _analyze_circuit(circuit: QuantumCircuit) -> tuple[list[_MeasureInfo], int]:
+    """Analyzes the information for each creg in a circuit."""
+    meas_info = []
+    max_num_bits = 0
+    for creg in circuit.cregs:
+        name = creg.name
+        num_bits = creg.size
+        if num_bits != 0:
+            start = circuit.find_bit(creg[0]).index
+        else:
+            start = 0
+        meas_info.append(
+            _MeasureInfo(
+                creg_name=name,
+                num_bits=num_bits,
+                num_bytes=_min_num_bytes(num_bits),
+                start=start,
+            )
+        )
+        max_num_bits = max(max_num_bits, start + num_bits)
+    return meas_info, _min_num_bytes(max_num_bits)
+
+
+def _prepare_memory(results: list[Result]) -> list[list[str]]:
+    """Joins splitted results if exceeding max_experiments"""
+    lst = []
+    for res in results:
+        for exp in res.results:
+            if hasattr(exp.data, "memory") and exp.data.memory:
+                lst.append(exp.data.memory)
+            else:
+                # no measure in a circuit
+                lst.append(["0x0"] * exp.shots)
+    return lst
+
+
+def _memory_array(results: list[list[str]], num_bytes: int) -> NDArray[np.uint8]:
+    """Converts the memory data into an array in an unpacked way."""
+    lst = []
+    for memory in results:
+        if num_bytes > 0:
+            data = b"".join(int(i, 16).to_bytes(num_bytes, "big") for i in memory)
+            data = np.frombuffer(data, dtype=np.uint8).reshape(-1, num_bytes)
+        else:
+            # no measure in a circuit
+            data = np.zeros((len(memory), num_bytes), dtype=np.uint8)
+        lst.append(data)
+    ary = np.asarray(lst)
+    return np.unpackbits(ary, axis=-1, bitorder="big")
+
+
+def _samples_to_packed_array(
+    samples: NDArray[np.uint8], num_bits: int, start: int
+) -> NDArray[np.uint8]:
+    """Converts an unpacked array of the memory data into a packed array."""
+    # samples of `Backend.run(memory=True)` will be the order of
+    # clbit_last, ..., clbit_1, clbit_0
+    # place samples in the order of clbit_start+num_bits-1, ..., clbit_start+1, clbit_start
+    if start == 0:
+        ary = samples[:, -start - num_bits :]
+    else:
+        ary = samples[:, -start - num_bits : -start]
+    # pad 0 in the left to align the number to be mod 8
+    # since np.packbits(bitorder='big') pads 0 to the right.
+    pad_size = -num_bits % 8
+    ary = np.pad(ary, ((0, 0), (pad_size, 0)), constant_values=0)
+    # pack bits in big endian order
+    ary = np.packbits(ary, axis=-1, bitorder="big")
+    return ary

qiskit/primitives/base/__init__.py

@@ -0,0 +1,20 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2022.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+
+"""
+Abstract base classes for primitives module.
+"""
+
+from .base_sampler import BaseSampler, BaseSamplerV1, BaseSamplerV2
+from .base_estimator import BaseEstimator, BaseEstimatorV1, BaseEstimatorV2
+from .estimator_result import EstimatorResult
+from .sampler_result import SamplerResult

qiskit/primitives/base/base_estimator.py

@@ -0,0 +1,252 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2022, 2023, 2024.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+
+"""Base Estimator V1 and V2 classes"""
+
+from __future__ import annotations
+
+from abc import abstractmethod, ABC
+from collections.abc import Iterable, Sequence
+from copy import copy
+from typing import Generic, TypeVar
+
+from qiskit.circuit import QuantumCircuit
+from qiskit.providers import JobV1 as Job
+from qiskit.quantum_info.operators import SparsePauliOp
+from qiskit.quantum_info.operators.base_operator import BaseOperator
+from qiskit.utils.deprecation import deprecate_func
+
+from ..containers import (
+    DataBin,
+    EstimatorPubLike,
+    PrimitiveResult,
+    PubResult,
+)
+from ..containers.estimator_pub import EstimatorPub
+from . import validation
+from .base_primitive import BasePrimitive
+from .base_primitive_job import BasePrimitiveJob
+
+T = TypeVar("T", bound=Job)
+
+
+class BaseEstimatorV1(BasePrimitive, Generic[T]):
+    r"""Estimator V1 base class.
+
+    Base class for Estimator that estimates expectation values of quantum circuits and observables.
+
+    An estimator is initialized with an empty parameter set. The estimator is used to
+    create a :class:`~qiskit.providers.JobV1`, via the
+    :meth:`qiskit.primitives.Estimator.run()` method. This method is called
+    with the following parameters
+
+    * quantum circuits (:math:`\psi_i(\theta)`): list of (parameterized) quantum circuits
+      (a list of :class:`~qiskit.circuit.QuantumCircuit` objects).
+
+    * observables (:math:`H_j`): a list of :class:`~qiskit.quantum_info.SparsePauliOp`
+      objects.
+
+    * parameter values (:math:`\theta_k`): list of sets of values
+      to be bound to the parameters of the quantum circuits
+      (list of list of float).
+
+    The method returns a :class:`~qiskit.providers.JobV1` object, calling
+    :meth:`qiskit.providers.JobV1.result()` yields the
+    a list of expectation values plus optional metadata like confidence intervals for
+    the estimation.
+
+    .. math::
+
+        \langle\psi_i(\theta_k)|H_j|\psi_i(\theta_k)\rangle
+
+    Here is an example of how the estimator is used.
+
+    .. code-block:: python
+
+        from qiskit.primitives import Estimator
+        from qiskit.circuit.library import RealAmplitudes
+        from qiskit.quantum_info import SparsePauliOp
+
+        psi1 = RealAmplitudes(num_qubits=2, reps=2)
+        psi2 = RealAmplitudes(num_qubits=2, reps=3)
+
+        H1 = SparsePauliOp.from_list([("II", 1), ("IZ", 2), ("XI", 3)])
+        H2 = SparsePauliOp.from_list([("IZ", 1)])
+        H3 = SparsePauliOp.from_list([("ZI", 1), ("ZZ", 1)])
+
+        theta1 = [0, 1, 1, 2, 3, 5]
+        theta2 = [0, 1, 1, 2, 3, 5, 8, 13]
+        theta3 = [1, 2, 3, 4, 5, 6]
+
+        estimator = Estimator()
+
+        # calculate [ <psi1(theta1)|H1|psi1(theta1)> ]
+        job = estimator.run([psi1], [H1], [theta1])
+        job_result = job.result() # It will block until the job finishes.
+        print(f"The primitive-job finished with result {job_result}"))
+
+        # calculate [ <psi1(theta1)|H1|psi1(theta1)>,
+        #             <psi2(theta2)|H2|psi2(theta2)>,
+        #             <psi1(theta3)|H3|psi1(theta3)> ]
+        job2 = estimator.run([psi1, psi2, psi1], [H1, H2, H3], [theta1, theta2, theta3])
+        job_result = job2.result()
+        print(f"The primitive-job finished with result {job_result}")
+    """
+
+    __hash__ = None
+
+    def __init__(
+        self,
+        *,
+        options: dict | None = None,
+    ):
+        """
+        Creating an instance of an Estimator V1, or using one in a ``with`` context opens a session that
+        holds resources until the instance is ``close()`` ed or the context is exited.
+
+        Args:
+            options: Default options.
+        """
+        super().__init__(options)
+
+    def run(
+        self,
+        circuits: Sequence[QuantumCircuit] | QuantumCircuit,
+        observables: Sequence[BaseOperator | str] | BaseOperator | str,
+        parameter_values: Sequence[Sequence[float]] | Sequence[float] | float | None = None,
+        **run_options,
+    ) -> T:
+        """Run the job of the estimation of expectation value(s).
+
+        ``circuits``, ``observables``, and ``parameter_values`` should have the same
+        length. The i-th element of the result is the expectation of observable
+
+        .. code-block:: python
+
+            obs = observables[i]
+
+        for the state prepared by
+
+        .. code-block:: python
+
+            circ = circuits[i]
+
+        with bound parameters
+
+        .. code-block:: python
+
+            values = parameter_values[i].
+
+        Args:
+            circuits: one or more circuit objects.
+            observables: one or more observable objects. Several formats are allowed;
+                importantly, ``str`` should follow the string representation format for
+                :class:`~qiskit.quantum_info.Pauli` objects.
+            parameter_values: concrete parameters to be bound.
+            run_options: runtime options used for circuit execution.
+
+        Returns:
+            The job object of EstimatorResult.
+
+        Raises:
+            TypeError: Invalid argument type given.
+            ValueError: Invalid argument values given.
+        """
+        # Validation
+        circuits, observables, parameter_values = validation._validate_estimator_args(
+            circuits, observables, parameter_values
+        )
+
+        # Options
+        run_opts = copy(self.options)
+        run_opts.update_options(**run_options)
+
+        return self._run(
+            circuits,
+            observables,
+            parameter_values,
+            **run_opts.__dict__,
+        )
+
+    @abstractmethod
+    def _run(
+        self,
+        circuits: tuple[QuantumCircuit, ...],
+        observables: tuple[SparsePauliOp, ...],
+        parameter_values: tuple[tuple[float, ...], ...],
+        **run_options,
+    ) -> T:
+        raise NotImplementedError("The subclass of BaseEstimator must implement `_run` method.")
+
+
+class BaseEstimator(BaseEstimatorV1[T]):
+    """DEPRECATED. Type alias for Estimator V1 base class.
+
+    See :class:`.BaseEstimatorV1` for details.
+    """
+
+    @deprecate_func(
+        since="1.2",
+        additional_msg="The `BaseEstimator` class is a type alias for the `BaseEstimatorV1` "
+        "interface that has been deprecated in favor of explicitly versioned interface classes. "
+        "It is recommended to migrate all implementations to use `BaseEstimatorV2`. "
+        "However, for implementations incompatible with `BaseEstimatorV2`, `BaseEstimator` can "
+        "be replaced with the explicitly versioned `BaseEstimatorV1` class.",
+    )
+    def __init__(
+        self,
+        *,
+        options: dict | None = None,
+    ):
+        """
+        Creating an instance of an Estimator, or using one in a ``with`` context opens a session that
+        holds resources until the instance is ``close()`` ed or the context is exited.
+
+        Args:
+            options: Default options.
+        """
+        super().__init__(options=options)
+
+
+class BaseEstimatorV2(ABC):
+    r"""Estimator V2 base class.
+
+    An estimator estimates expectation values for provided quantum circuit and
+    observable combinations.
+
+    An Estimator implementation must treat the :meth:`.run` method ``precision=None``
+    kwarg as using a default ``precision`` value.  The default value and methods to
+    set it can be determined by the Estimator implementor.
+    """
+
+    @staticmethod
+    def _make_data_bin(_: EstimatorPub) -> type[DataBin]:
+        # this method is present for backwards compat. new primitive implementations
+        # should avoid it.
+        return DataBin
+
+    @abstractmethod
+    def run(
+        self, pubs: Iterable[EstimatorPubLike], *, precision: float | None = None
+    ) -> BasePrimitiveJob[PrimitiveResult[PubResult]]:
+        """Estimate expectation values for each provided pub (Primitive Unified Bloc).
+
+        Args:
+            pubs: An iterable of pub-like objects, such as tuples ``(circuit, observables)``
+                  or ``(circuit, observables, parameter_values)``.
+            precision: The target precision for expectation value estimates of each
+                       run Estimator Pub that does not specify its own precision. If None
+                       the estimator's default precision value will be used.
+
+        Returns:
+            A job object that contains results.
+        """

qiskit/primitives/base/base_primitive.py

@@ -0,0 +1,45 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2022.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+
+"""Primitive V1 abstract base class."""
+
+from __future__ import annotations
+
+from abc import ABC
+
+from qiskit.providers import Options
+
+
+class BasePrimitive(ABC):
+    """Primitive V1 abstract base class."""
+
+    def __init__(self, options: dict | None = None):
+        self._run_options = Options()
+        if options is not None:
+            self._run_options.update_options(**options)
+
+    @property
+    def options(self) -> Options:
+        """Return options values for the estimator.
+
+        Returns:
+            options
+        """
+        return self._run_options
+
+    def set_options(self, **fields):
+        """Set options values for the estimator.
+
+        Args:
+            **fields: The fields to update the options
+        """
+        self._run_options.update_options(**fields)

qiskit/primitives/base/base_primitive_job.py

@@ -0,0 +1,78 @@
+# This code is part of Qiskit.
+#
+# (C) Copyright IBM 2024.
+#
+# This code is licensed under the Apache License, Version 2.0. You may
+# obtain a copy of this license in the LICENSE.txt file in the root directory
+# of this source tree or at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# Any modifications or derivative works of this code must retain this
+# copyright notice, and modified files need to carry a notice indicating
+# that they have been altered from the originals.
+"""
+Primitive job abstract base class
+"""
+
+from abc import ABC, abstractmethod
+from typing import Generic, TypeVar, Union
+
+from ..containers import PrimitiveResult
+from .base_result import _BasePrimitiveResult
+
+ResultT = TypeVar("ResultT", bound=Union[_BasePrimitiveResult, PrimitiveResult])
+StatusT = TypeVar("StatusT")
+
+
+class BasePrimitiveJob(ABC, Generic[ResultT, StatusT]):
+    """Primitive job abstract base class."""
+
+    def __init__(self, job_id: str, **kwargs) -> None:
+        """Initializes the primitive job.
+
+        Args:
+            job_id: A unique id in the context of the primitive used to run the job.
+            kwargs: Any key value metadata to associate with this job.
+        """
+        self._job_id = job_id
+        self.metadata = kwargs
+
+    def job_id(self) -> str:
+        """Return a unique id identifying the job."""
+        return self._job_id
+
+    @abstractmethod
+    def result(self) -> ResultT:
+        """Return the results of the job."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `result` method.")
+
+    @abstractmethod
+    def status(self) -> StatusT:
+        """Return the status of the job."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `status` method.")
+
+    @abstractmethod
+    def done(self) -> bool:
+        """Return whether the job has successfully run."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `done` method.")
+
+    @abstractmethod
+    def running(self) -> bool:
+        """Return whether the job is actively running."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `running` method.")
+
+    @abstractmethod
+    def cancelled(self) -> bool:
+        """Return whether the job has been cancelled."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `cancelled` method.")
+
+    @abstractmethod
+    def in_final_state(self) -> bool:
+        """Return whether the job is in a final job state such as ``DONE`` or ``ERROR``."""
+        raise NotImplementedError(
+            "Subclass of BasePrimitiveJob must implement `is_final_state` method."
+        )
+
+    @abstractmethod
+    def cancel(self):
+        """Attempt to cancel the job."""
+        raise NotImplementedError("Subclass of BasePrimitiveJob must implement `cancel` method.")